Discourse Completion Tasks (DCTs)

Since it was first adopted in Cross-cultural Speech Act Realization Project (CCSARP) (Blum-Kulka, 1982), Discourse Completion Tasks (DCTs) has been the most widely used device in studies on cross-cultural and interlanguage speech act realizations (e.g. Blum-Kulka, House & Kasper, 1989;Blum-Kulka & Olshtain, 1984, 1986; Barron, 2003;Le Pair, 1996).

In recent studies, various item formats of DCT are developed (Kasper & Rose, 2002) and the task effect on participants’ choices of strategies are examined (Bardovi-Harling & Hartford, 1993; Billmyer & Varghese, 2000). On task format effects, one of the controversial issues is the influence of a rejoinder of a DCT item on participant’s choices of strategies, which is not discussed in CCSARP though it was first applied in the study. Rose (1992) used two forms of DCT—one with rejoinders and one without rejoinders— to investigate whether the request act of EL1 participants would vary or not. The result indicated that participants’ responses of request were longer in DCT items without rejoinders than those in DCT items with

rejoinders though the difference was not statistically significant. In Johnston, Kasper &

Rose’s (1998) study on the effect of rejoinders on participants’ speech acts of complaint, request and apology, they found that participants’ use of strategies and modification varied according to two dimensions: whether rejoinders exist or not and the types of rejoinders, that is, negative (non-compliance) or positive (compliance) ones. For instance, the result of their study on request showed that when the rejoinder was negative, non-native speaker preferred internal modifies to external modifications as opposed to native speakers of English. Thus, Johnston et al (1998) proposed that data collected by DCT with rejoinders may not be comparable directly. To elicit cross-culturally comparable data, DCT without rejoinders is preferred in recent pragmatic studies (Byon, 2004; Hill, 1997; Takahashi & Beebe, 1993, Yu, 1999) as illustrated below.

You are going to visit your friend, who lives in the college dormitory. You are on campus, but don’t know where the dorm is. You are going to ask a student,

who is passing by, for the location of the dorm. How will you ask the student?

(Adapted from Byon, 2004:1701)

The validity of DCT has been examined by some researchers. Hinkel (1997) criticized DCT that participants’ spoken utterances were elicited by written forms. However, in Beebe and Cummings’ (1996) study on methods of data collection on refusals, they compared DCT with naturalistic data and the result revealed that the initial classification of semantic formulas of these two methods were identical though some differences such as actual wording used, the length of responses, etc. were found. The same result was reported by Eisenstein & Bodman (1993) that participants showed similar semantic strategies on tasks of DCT, natural observation, oral questionnaires, and open role-plays. Furthermore, in Kasper & Dahl’s

(1991) review of interlanguage studies, DCT was used in 19 out of 35 studies and the advantages have been advocated by most of researchers. First, DCT is not only a practical data-gathering instrument that makes it possible to collect large scale of data effectively in a short period of time (Beebe & Cummings, 1996; Johnston, Kasper & Rose, 1998) but also an appropriate means to obtain more stereotyped formulas and strategies in everyday speech acts (Barron, 2003; Blum-Kulka et al, 1989). Second, situational variables such as social status, distance, gender, the degree of imposition etc., can be effectively controlled so that it enables researchers to focus on a given language phenomenon and further investigate, predict and test it comparably.(Blum-Kulka et al, 1989;Johnston et al, 1998). Take two scenarios designed for the present study as shown in Table 3.1 for example, to examine the effect of imposition on participants’ choices of strategies, comparable data could be obtained in situation 1 and 11 of the questionnaire since Status and Interlocutors’ Gender were designated as constant variables, the degree of imposition as the independent variable and the observation of the request utterances as the dependent variables. As what Bardovi-Harlig (1999: 238) proposed, DCT is “…the most celebrated and most maligned of all the methods used in cross-cultural and interlanguage pragmatics research” and instruments per se are not either good or bad, but only appropriate or not appropriate for the purpose of the study.

Table 3. 1The Distribution of factors in situation 1 and 11 Variable

Situation Status Interlocutor’s

Gender Distance Imposition

1 ┼ f ┼

11 ┼ F －

Variables that affect linguistic politeness of the realization of the request act include social distance, social status, rank of imposition (Brown & Levinson, 1987), right and obligation to make requests (Blum-Kulka & House, 1989:131; Trosborg, 1995). The request here was defined as the Right-type request that the requester has no right to require the compliance and the requestee has no obligation to perform the demanded acts.

Obligation-type request was ignored since it was irrelevant with FTA (Liao, 1997) which was the main issue in the present study.

The DCT applied in the present study consisted of 20 request-eliciting scenarios in which the four variables were to be examined—social distance, social status, imposition and interlocutors’ gender, which are critical factors that influence participants’ requests strategies in both Chinese and Western cultures (Brown & Levinson, 1987, Hong, 1996; Liao &

Bresnahan, 1996; Yu, 1999). Situations designed were confined in the university setting in order to control the effect of extraneous variables such as age, social economic status, etc.

The first variable” social distance” is defined along the scale of the frequency of interaction or the familiarity between interlocutors (Brown & Levinson, 1987) and is conventionally categorized into three levels in sociopragmatic studies—stranger, acquaintance and intimate (Holmes, 1992,1995; Leech, 1983;Wolfson, 1988). To further eliminate the factor of personal family relationship, family members were excluded as intimate in the study. Thus, roles represented the three levels were total strangers, classmates, and close friends accordingly in the university setting,

The second variable” social status” in the Western culture is defined as a relative power or hierarchical status that one has the authority (Leech, 1983) or ability to impose his/her will, evaluations over another (Brown & Levinson, 1987: 77) while in Chinese context, it can be

determined according to two parameters respectively: power, same as Brown& Levinson’s, and age (Lee-Wong, 1994). To elicit the comparable data, Brow& Levinson’s definition of social status was adopted and categorized as the conventional 3 levels: high, equal, and low.

Roles represented the three status levels in university setting were professors, classmates, and service people, respectively.

The third variable “ranking of imposition” is determined by (1) services such as time or efforts that have to be spent on carrying out certain speech acts (2) goods (including material goods such as money, pen, etc. and non-material goods such as information or other face payments) (3) the obligation and right to perform the required act (Brown & Levinson, 1987).

The rank order of impositions is conventionally classified into binary levels: high and low.

For instance, the imposition of borrowing USD100 dollars is higher than that of borrowing 1 dollar.

The fourth factor was Interlocutors’ Gender. According to Zuengler (1994), second learners’ communication strategies were influenced by interlocutors’ gender whereas the impact of participants’ gender was not obvious. Billmyer (1993) also pointed out that speakers seemed to be aware of interlocutors’ gender and made accommodation of speech features to the interlocutors in conversations. Thus, the factor Interlocutors’ Gender was also concerned when the questionnaire was designed.

The scenarios were designated for each of the three levels of “social status” and “social distance and these two variables resulted in 5 scenarios because one overlapped level –the mid level (social status equal and social distance equal)—was deducted. When the other two-level variables, gender and imposition, were taken into consideration, the design were 5 (status+ distance)*2(gender)*2 (imposition) and scenarios amounted to 20 in total as showed

in Table 3.1 below. Each scenario consisted of two parts (3.1): Part A Scaled-response questionnaire and Part B DCT, which was to elicit participants’ sociolinguistic and pragmalingustic responses respectively as illustrated below. There were two versions of questionnaires: English and Chinese. The English version was proofread by a native speaker of English who is also familiar with the Chinese cultural so as to ensure that the situations were cross-culturally equivalent.

(3.1) An example of SRQ and DCT used in the present study

【Situation1】Ms. Vivian Rupert is a professor in your department. You are one of the members of the students’ organization and are responsible for raising funds for an activity that will be held by the organization next month. She is in her office now and you want to knock on the door and ask her to sponsor your organization by donating for USD 160 dollars.

Part A. Complete the following (1) ~ (3) by putting “X” in the bracket, 1 being small, 5 being big.

(1) How great an imposition do you think the requested act will be on Ms.

Vivian Rupert?

［］1 ［］2 ［］3 ［］4 ［］5

(2) How difficult do you think it would be for you to make such a request?

［］1 ［］2 ［］3 ［］4 ［］5

(3) What is the likelihood that you will make a request in this situation?

［］1 ［］2 ［］3 ［］4 ［］5

Part B. What would you say in this situation? Please write a response specifically.

I would say (to Ms. Vivian Rupert):

_______________________________________________________

_______________________________________________________

Table 3. 2The distribution of variables in 20 situations of the questionnaire Item Imposition Gender Status Distance Situations

1 ＋ Ask a professor to sponsor NT5000 for an

activity of the Department.

17 = = Ask a classmate to lend you an expensive

digital camera for one more week.

8 － Ask your close friend to make presentation of

－ Ask a waitress to change your order which is preparing because you felt you would like to try another dish for certain reasons.

－ Ask a club member to stay for rehearsal though it’s 10.p.m. mobile phone to make a call with pay.

10 you make two pieces of copies first.

20

－

M

－ Ask a junior team member to find another member who has not shown up.

Note: +: high; =: equal; –: low

Status / Distance +: hearers’ (H) status > speakers’ (S) Status / Distance =: H’ status =H

Status / Distance –: H’ status < S

F / M: Female interlocutors; Male interlocutors